Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Estimation of Dermal Exposure to Oil Spill Response and Clean-up Workers after the Deepwater Horizon Disaster

Estimation of Dermal Exposure to Oil Spill Response and Clean-up Workers after the Deepwater... The GuLF STUDY is investigating health outcomes associated with oil spill-related chemical ex- posures among workers involved in the spill response and clean-up following the Deepwater Horizon disaster. Due to the lack of dermal exposure measurements, we estimated dermal ex- posures using a deterministic model, which we customized from a previously published model. Workers provided information on the frequency of contact with oil, tar, chemical dispersants ap- plied to the oil spill and sea water, as well as the use of protective equipment, by job/activity/ task. Professional judgment by industrial hygienists served as a source of information for other model variables. The model estimated dermal exposures to total hydrocarbons (THC), benzene, ethylbenzene, toluene, xylene, n-hexane (BTEX-H), polycyclic aromatic hydrocarbons (PAHs), and dispersants in GuLF DREAM units (GDUs). Arithmetic means (AMs) of THC exposure estimates across study participants ranged from <0.02 to 5.50 GDUs for oil and <0.02 to 142.14 GDUs for tar. Statistical differences in the estimates were observed among the AMs of the estimates for some broad groups of worker activities over time and for some time periods across the broad Published by Oxford University Press on behalf of The British Occupational Hygiene Society 2021. Downloaded from https://academic.oup.com/annweh/article/66/Supplement_1/i234/6395048 by DeepDyve user on 18 July 2022 Annals of Work Exposures and Health, 2022, Vol. 66, No. S1 i235 groups of activities. N-Hexane had ranges similar to THC for oil exposures (e.g. AMs up to 2.22 GDUs) but not for tar (up to 5.56 GDUs). Benzene, ethylbenzene, toluene, and xylene, in contrast, were characterized by higher exposure levels than THC for oil (AMs up to 12.77, 12.17, 17.45, and 36.77 GDUs, respectively) but lower levels than THC to tar (AMs up to 3.69, 11.65, 42.37, and 88.18 GDUs, respectively). For PAHs, the AMs were as high as 219.31 and 587.98 for oil and tar, re- spectively. Correlations of these seven substances to each other were high (>0.9) for most of the substances in oil but were lower for some of the substances in tar. These data were linked to the study participants to allow investigation of adverse health effects that may be related to dermal exposures. Keywords: Deepwater Horizon; dermal exposure, exposure assessment; oil spill; total hydrocarbons Introduction This paper describes the methods and the results for the dermal assessments. On 20 April 2010, the Deepwater Horizon oil rig ex- An overview of the exposure assessment effort ploded in the Gulf of Mexico, causing almost 5 million for the STUDY is presented in Stewart et  al. (2021). barrels of oil to be released into the Gulf waters over the Development of exposure groups (EGs) is described in following 3 months. Over 55 000 workers were rostered Stenzel, Arnold et al., 2021. The assessment of airborne by NIOSH as having participated in the response and exposures to THC and BTEX-H is described in Huynh clean-up (NIOSH, 2011). Workers had inhalation and et al., 2021a,b,c; Ramachandran et  al., 2021; Groth, dermal exposures to multiple oil-related compounds, as Banerjee et al., 2021; and Groth, Huynh et al., 2021. well as possible exposure to chemical dispersants, PM , 2.5 Assessment of other airborne exposures is also reported and cleaning products. Although more than 160 000 air (PM (Pratt et al., 2021); dispersant aerosols (Arnold 2.5 measurements were available to characterize inhalation et al., 2021) and vapors (Stenzel, Groth et al., 2021); and exposures, no dermal or surface wipe measurements oil mists (Stewart et al., 2021)). had been collected. Furthermore, few measurements of dermal exposure were available from other spills for ex- posure characterization. Background The Gulf Long-term Follow-up Study (GuLF The Deepwater Horizon (DWH) oil spill led to a mas- STUDY), initiated by the National Institute of sive effort to contain the spill and clean the Gulf of Environmental Health Sciences (NIEHS), is Mexico waters and shoreline. Most of the OSRC ac- investigating potential adverse health effects associated tivities were suspected as having resulted in dermal with the oil spill response and clean-up (OSRC) (Kwok exposure to oil, oily salt water, and tar. Two rigs (the et  al., 2017). As part of the exposure assessment ef- Discoverer Enterprise (Enterprise) and the Helix fort, we updated and enhanced a previously published Q4000 (Q4000)), were involved in mitigating the re- dermal exposure deterministic model (Van Wendel de lease, capturing the leaking oil/natural gas mixture; Joode et al., 2003) to better reflect the contribution of and separating the gas from the oil and flaring the gas various exposure determinants relevant to the GuLF (Enterprise) or the oil/gas mixture (Q4000) Huynh et al., STUDY (Gorman Ng et al., 2021). Using information 2021a. These rigs were located within 1 nautical mile from both the study participants and the study indus- (nmi, 1.8 km; 1.1 mi) of the wellhead, approximately trial hygienists, we estimated exposure to total hydro- 50 nmi (93 km) southeast of the Louisiana (LA) shore. carbons (THC), benzene, toluene, ethylbenzene, xylene, Two other drilling rigs, the Development Driller II n-hexane (BTEX-H), polycyclic aromatic hydrocarbons (DDII) and the Development Driller III (DDIII), lo- (PAHs) (as a single substance), and (total) dispersants cated within the 1 nmi radius of the wellhead (referred due to these substances’ inhalation toxicity and their to as the hot zone), were each responsible for drilling a ability to be absorbed into the skin or adversely affect relief well. the skin. (Dispersant refers to chemicals sprayed onto Supporting these four vessels was a sizable, but un- an oil slick on the water surface or injected into the known, number of other large marine vessels (MVs) water to break down the oil into small droplets that (Huynh et al., 2021c; Ramachandran et  al., 2021). more readily mix with the water. It is the dispersants’ Fourteen MVs that piloted remotely operated vehicles components that are associated with possible toxicity.) Downloaded from https://academic.oup.com/annweh/article/66/Supplement_1/i234/6395048 by DeepDyve user on 18 July 2022 i236 Annals of Work Exposures and Health, 2022, Vol. 66, No. S1 (ROVs), called here ROV vessels, performed several Methods underwater activities, such as moving equipment, col- After review of several dermal assessment models, the lecting water samples and taking videos. Other MVs DREAM model (Van Wendel de Joode B et al., 2003) provided other types of support, for example, pumping was selected as the most appropriate model for the fluids into the well for well closure attempts; spraying GuLF STUDY because it allowed the use of information water onto the flaring vessels to reduce temperatures; on determinants of dermal exposure, some of which storing and transporting collected oil; supplying ma- had been collected from study participants through a terials/chemicals/crew; and spraying dispersant onto telephone interview. Validation work conducted by the the water’s surface near the rigs. We called the 5 nmi (9 developers of the original model found that the dermal km) radius around the wellhead, excluding the hot zone, exposure units (DREAM units, a dimensionless unit) where most of the supporting vessels worked, the source. correlated well with hand exposure measurements Research vessels collected samples of water and across a range of work sites and exposure agents, but oil, took water quality measurements, monitored the less well with other body parts (Van Wendel de Joode B oil plume and collected oiled and dead wildlife. Other, et al., 2005). typically smaller, vessels, such as fishing and shrimping We updated the DREAM model to reflect more recent boats, scouted for oil; deployed/maintained/retrieved studies and adapted it, primarily by changing weights of booms; skimmed or burned oil off the water surface; the various determinants to fit the oil-related chemicals collected wildlife; carried personnel, equipment and and dispersants and available data of the GuLF STUDY supplies to and from the wellhead area; and transported to allow retrospective evaluation using interview data collected oil and oily water back to shore Huynh et al., from study participants (Gorman Ng et al., 2021). We 2021b. Vessels also decontaminated (deconned) other called this model GuLF DREAM and the dimensionless vessels, jetties and other manmade structures using pres- outcome measures, GuLF DREAM units (GDUs). The sure spraying. Operations occurred throughout the Gulf model required variables for: of Mexico, north of the wellhead off the LA, Mississippi (MS), Alabama (AL), and Florida (FL) coastlines. • the concentration, vapor pressure score, and viscosity Activities on land (Huynh et al., 2021c) included score of each substance; loading/unloading of everything related to the vessels’ • each of three exposure pathways: emission (direct functions described above at ports and docks, primarily contact with the liquid substance), deposition (con- in LA, MS, AL, and FL. Deconning of vessels, equipment tact with the airborne substance), and surface on the vessels, boom and other gear was done by low- or transfer (contact with a contaminated surface); high-pressure spraying or by hand (e.g. rags or absorb- • the surface area of each of nine body parts (head, ents). Beaches and marshes were systematically patrolled upper arm, lower arm, hands, front of torso, back of to evaluate locations needing attention, after which they torso, upper legs, lower legs, and feet); were cleaned by picking up oil; oily water; tar balls, • for each body part, the intensity (i.e. the percentage patties, and mousse; oiled plants; and garbage by hand or of surface area covered) and the frequency of ex- using handheld equipment for disposal in bags. Wildlife posure for the likeliest pathway for oil and tar; was captured and rehabilitated. Support staff included ma- • water (seawater removes water soluble compounds) terial handling, cooks, housekeeping, office, and security contact; and workers. • the type and frequency of replacement of gloves and As the oil was released from the well, it changed com- protective clothing. position, due to natural processes including evaporation, Each variable was assigned a value, generally between 0 dispersion, emulsification, dissolution, photo-oxidation, and 5 (Gorman Ng et al., 2021). sedimentation, and biodegradation. These processes oc- To evaluate our model, exposures measured in two curred from the time the oil was released in the subsea studies of heavy fuel oil (in a variety of industries unre- water and continued while it was on the water surface lated to oil spills (Christopher et al., 2007)) and of as- and on land. This change in oil, called weathering, re- phalt (among paving workers (Cavallari et al., 2012)) sulted over time in a differential decrease of the percent- were estimated using GuLF DREAM and compared to ages of the volatile chemicals and a differential increase those studies’ measurements. A correlation (ρ) of 0.59 in the percentages of the semi- and non-volatile chem- was found between the GuLF DREAM estimates and the icals in the oils and tars (Stenzel, Arnold et al., 2021). measurements for the hands (Gorman Ng et al., 2021). Downloaded from https://academic.oup.com/annweh/article/66/Supplement_1/i234/6395048 by DeepDyve user on 18 July 2022 Annals of Work Exposures and Health, 2022, Vol. 66, No. S1 i237 There were insufficient measurements to evaluate other Questions on the participant’s use of protective equip- body parts. ment by job/activity/task covered the use of: Because dermal exposure was likely to have occurred • leather, cotton, or synthetic gloves; and if a second both to oil and to tar, we assessed a variety of chemicals glove was worn; contained in each of those two substances, i.e. THC, as • boots or rubber slip-ons; petroleum hydrocarbons; each of the BTEX-H chemicals; • protective coveralls such as Tyvek; and PAHs as a single substance. For dispersants, based on • long sleeved shirts, jackets, or coveralls; and the relative quantity of each dispersant used, we evaluated • the frequency of changing each of these types of a 33.4/66.6 mixture of COREXIT™ 9527A and 9500A clothing. (based on the amounts of the two dispersants used) applied by air across the Gulf and 9500A alone at the wellhead, ei- All questions were asked of all study participants, except ther sprayed on the water or injected near the seabed. for the set of questions on the frequency of exposure to an unspecified chemical for the eight body parts, which was Data collection only asked of the home interview participants. See the on- In the GuLF STUDY telephone interview, participants line Supplemental Material (SM), Table S1 for more details. (N = 24 937) were asked about the variety of jobs/activ- Processing of information for the estimates ities/tasks presented in Background, as well as questions that addressed dermal exposure specific to most of the Data coding A set of rules was developed to translate the question re- reported job/activities/tasks (for exceptions, see below). A subset of the participants (N = 11 193) later com- sponses into weights for the GuLF DREAM model (Gorman Ng et al., 2021). The same weights were assigned for oil, pleted a home visit interview that included other dermal questions. The dermal questions asked were: tar and dispersants. For example, if a participant said yes to “Did your skin or clothing come in contact with (substance) • for each activity reported by the participant, if his/ during any of your oil spill clean-up work?”, a value of 1 was her skin or clothing had contact with each of several assigned if the response was yes and 0 if the response was substances, including oil, tar, dispersants, and water. no. If yes, a follow-up question was asked, “On an average (We were unable to find a definition for the difference workday, how much of the time was your skin or clothing in between oil and tar that we thought the participants contact with (substance)?” Responses were coded on a scale would be able to distinguish. We therefore defined of 1–5: “None” = 1, “<1/2” = 2, “About ½” = 3, “>1/2” = 4, the two substances as “a solid or gooey oily residue and “All of it” = 5. Rules were developed based on the sub- or tar” (asked first) and “oil or oily water”). stance being assessed, the question number, and the variable. Questions were missing responses for approximately If contact occurred, additional questions were asked for 5–10% of participants. In addition, only approximately each reported job/activity/task: one-third of the cohort, i.e. those who had home visits, • the frequency of the contact (none; less than half the had responses on frequency of contact to the eight body time; about half; more than half; or all the time); parts. To impute information from both types of missing • whether the substance got on the hands (yes/no); data, a job-exposure matrix (JEM) approach was taken. • the number of hours a day it was on the hands before The basis for the JEM was the set of EGs, unique com- being washed off (h, min); binations of jobs/activities/tasks, location, and time • whether the substances got on the skin or clothing (Stenzel, Arnold et al., 2021), developed for inhalation other than the hands (yes/no); exposures. Time periods (TPs) are described in the SM, • frequency of contact with water (none; less than Table S2 and jobs/activities/tasks in the SM, Table S3. half the time; about half; more than half; or all the Briefly, the events in each time period were: time); and • the frequency that each of the eight other body parts • TP1a (22 April through 14 May 2010): oil flowed became wet with a (unspecified) chemical (<1 day from the damaged well. Drilling started on a relief per month; 1–4 days per month; 1–5 days per week; well. Water clean-up activities started. Oil reached the or almost every day), whether on clothing or on the LA shoreline, and beach cleanup started. Dispersant skin. application began. Downloaded from https://academic.oup.com/annweh/article/66/Supplement_1/i234/6395048 by DeepDyve user on 18 July 2022 i238 Annals of Work Exposures and Health, 2022, Vol. 66, No. S1 • TP1b (15 May through 15 July 2010): oil flow the assigned JEM value was 1 (yes), and 0 (no) if <50% continued. Drilling began on a second relief well. of participants answered “yes”). Median integer scores Dispersant operations continued. Water clean-up ac- were used for categorical or continuous responses (e.g. tivities continued. The well was successfully mechan- for the question, “On an average workday, how much of ically capped on July 15, which essentially stopped the time was your skin or clothing in contact with oil?”, the release of oil. Beach and wildlife clean-up was the responses were coded as: “None” = 1, “<1/2” = 2, being done in all four states. “About ½” = 3, “>1/2” = 4, and “All of it” = 5. The in- • TP2 (16 July through 10 August 2010): the well was teger of the median score across all respondents was as- “static killed” on 10 August. Water activities started signed as the JEM value. diminishing. Beach and jetty clean-up and decontam- To minimize participant burden, questions were not ination of vessels and equipment continued. asked about changes over time. The JEM values for a • TP3 (11 August through 30 September 2010): given job/activity/task/location, however, varied based large-scale final decontamination of the vessels and on the responses of the participants who worked in related equipment started. Water activities continued each time period. For example, workers who reported to lessen. Beach clean-up continued but started to cleaning oil pools in LA in TP1a (N = 59) had a median decline. emission frequency of 3 (“1/2 the time”), whereas for • TP4 (1 October through 31 December 2010): water those who cleaned oil pools in TP1b (N = 461), the me- efforts essentially were completed by the end of dian emission frequency was 1 (“never”), although many December. Beach and marsh clean-up continued by of the 59 in TP1a also worked in TP1b. When <5 partici- decreasing numbers of people. pants or <10% of the participants provided responses to • TP5 (1 January to 31 March 2011)  and TP6 (1 an EG, the reported summary value was overridden and April to 30 June 2011): beach clean-up continued by replaced by the summary value for “All states” for that decreasing numbers of people. Water operations were activity and time period. limited to the near shore transfer of equipment, sup- Some GuLF DREAM variables were not sought plies, collected materials and personnel. TP6 is distin- from the study participants because the variables re- guished from TP5 because of the warmer ambient air quired information the participants were unlikely to temperatures. know. Instead, the study industrial hygienists supplied the information based on the scientific literature, know- Most of the inhalation EGs were retained, although ledge of exposures, extensive published and unpublished we modified some to incorporate additional informa- documentation on the DWH event, and consensus of the tion obtained from the interviews by adding groups to study hygienists. This information included three prop- account for differences relevant to dermal exposures. erties of the substances distinguished by the degree of For example, we had an inhalation EG for “Handled/ weathering: i.e. the concentration of the various com- cleaned wildlife”, which was as precise as the air meas- ponents in the oil or tar; a vapor pressure score; and a urement data allowed. For the dermal assessment, viscosity score. The values for these scores and their der- however, we distinguished among the various wildlife ivation are described in SM, Table S4. We also identified handling/cleaning activities, i.e. “Handled oily wild- the percentage of oil weathering associated with each life”, “Cleaned wildlife”, “Used soaps to clean wildlife” EG (SM, Table S5). and “Retrieved dead wildlife”. New EGs were also de- We assigned other exposure determinants. First, we veloped for workers on the rig vessels, because the job- assigned the primary pathway of exposure (emission or based inhalation EGs had small numbers of respondents. surface transfer) based on knowledge and photographs Small sample sizes would have resulted in less precise of the job/activity/task. We considered that the contribu- values when using study participants’ responses to im- tion of deposition was very small in light of the airborne pute missing data, so we combined the various jobs to concentrations measured relative to the contribution form 7 broad “jobs”. Descriptions of all jobs/activities/ from emission or surface transfer (the average of the tasks are provided in SM, Table S3. THC air vapor concentrations across EGs (Huynh et al., The data used to complete the JEM cells were sum- 2021a,b,c); Ramachandran et al., 2021 was 0.8 ppm). maries of the responses to each dermal-related question Thus, deposition was not considered a pathway and de- for each EG. Percentages were used for imputing yes/ position frequency and intensity were assigned “0”. no questions (e.g. if ≥50% of participants answered Second, the study industrial hygienists entered values “yes” to “Did your skin or clothing come in contact for the intensity received on each body part for the ap- with oil during any of your oil spill clean-up work?”, propriate pathway, i.e. emission or surface transfer. Downloaded from https://academic.oup.com/annweh/article/66/Supplement_1/i234/6395048 by DeepDyve user on 18 July 2022 Annals of Work Exposures and Health, 2022, Vol. 66, No. S1 i239 The DREAM definition of intensity was the amount of under weathering conditions of 25–30%, we combined the body part exposed, and we retained that definition the oil and tar estimates for any participant’s job/activity/ for GuLF DREAM. Intensity values and weights were: task, location and time period associated with those de- <10% of body part  =  1; 10–50% of body part  =  3; grees of weathering. This procedure primarily affected the ≥50% of body part = 10. A single intensity weight across latter time periods (TP3–6). Statistical differences in the all time periods was assigned to each body part for each AMs were identified by non-overlapping 95% confidence job/activity/task/location. intervals (upper confidence level, UCL; lower confidence Third, some types of clothing and wind speed values level, LCL). Pearson correlations were calculated among were assigned by the industrial hygienists. We assumed the various components of oil and of tar. headgear was cotton hats. We assumed that workers in all jobs/activities/tasks had been required to wear heavy (leather, cotton, or synthetic) gloves, except for security, Results general environment/land, housekeeping, kitchen, and The range of THC dermal AM estimates from oil ex- office workers, all of whom were assumed to have worn posure was broad, ranging from AMs <0.02 GDUs light (latex) gloves. Frequency of hat replacement was (among several groups, for example, “IH/safety-water”, considered less than daily and clothes and shoes (rubber “All states”, TP1a) to 5.50 GDUs (“Deconned booms/ booties), daily. The frequency of glove replacement, how- land”, “LA”, TP3) (not shown). Equivalent tar AMs were ever, was provided by the study participant: “less than <0.02 GDUs (such as “Ran mechanical equipment/ports daily”, “daily” or “within a work shift”. Wind speed was & docks” All states, TP3) to 142.14 GDUs (“Retrieving coded as 1 (i.e. no effect on the substance), except for boom in shallow water”, MS, TP6). dispersant, which was assigned 0.75 (small effect). For THC, no statistical differences in oil exposures Finally, participants were not asked about dermal occurred across time periods among the workers on exposures for jobs/activities/tasks that were expected to the rigs (Fig. 1 (note that the scales differ by substance) have no or very little dermal exposure, such as “Cooks”, and SM, Table S6) The values for the variables asso- “Office workers”, and “Security”. Other workers were ciated with the ROV, burner fire control and research linked to an EG from a response to an open-ended ques- vessels were assigned by the industrial hygienists and tion (“What else did you do?”) and therefore did not get therefore there was little variability in the estimates. asked dermal questions (e.g. workers on ROV and re- Other water workers were characterized by increasingly search vessels). The study industrial hygienists entered a higher (and statistically significant) mean dermal esti- single value for each variable across all states and time mates over time: AM  = 0.39 GDU (LCL: 0.38, UCL: TP1b periods to allow prediction for these jobs/activities/tasks. 0.39 GDU); AM   =  0.49 GDU (0.48, 0.50 GDU); TP2 Once all participants had complete information for AM  = 0.56 GDU (0.56, 0.57 GDU); AM   =  0.85 TP3 TP4 each job/activity/task, either from the responses or the GDU (0.83, 0.88 GDU); and AM  = 1.31 GDU (0.95, TP5 imputed data, we applied the GuLF DREAM model 1.83 GDU). For land workers, the AMs significantly fell to develop exposure estimates for each participant’s from TP (AM = 0.97 GDU (0.92, 1.01 GDU)) to TP 1a 1b unique combination of job/activity/task, location and and TP (AM for both time periods = 1.59 GDU (0.58, time period. Because multiple jobs/activities/tasks were 0.60 GDU)) and then started rising, with statistically reported (median=6 per participant), multiple estimates significant higher AMs in each of the later time periods were assigned per time period. (AM  = 0.92 GDU (0.91, 0.93 GDU); AM  = 1.16 TP3 TP4 GDU (1.14, 1.18 GDU); AM  = 1.32 GDU (1.27, 1.38 TP5 Statistical analyses GDU); and AM  = 1.67 GDU (1.56, 1.78 GDU). TP6 For presentation purposes, the estimated arithmetic means Tar exposure was assessed only for “Other water op- (AMs) of the GDUs across all study participants were erations” and “Land”, as the other vessels (rigs, ROVs, calculated by job/activity/task, location, time period and burner fire control vessels and RVs) had left the area substance. In the AM calculation, we dropped all job/ac- by the time the oil had likely weathered to tar. Water tivity/task, location, time period combinations that were workers were exposed to higher statistically signifi- <0.01 GDUs. For the graphs, we used the broad group- cant differences over time (AM  = 0.37 GDU (0.35, TP3 ings of workers for “All rigs”, “All ROVs”, “All research 0.38 GDU); AM   =  0.86 GDU (0.83, 0.89 GDU); TP4 vessels”, “Burner fire control vessels”, “Other water”, and AM  = 1.13 GDU (1.05, 1.21 GDU); and AM  = 1.46 TP5 TP6 “Land” by time period for “All states” for oil and for tar. GDU (1.31, 1.62 GDU)) (Fig. 2 and SM, Table S6). Due to our inability to distinguish between oil and tar Land workers’ tar exposures fell from TP1a to TP3 Downloaded from https://academic.oup.com/annweh/article/66/Supplement_1/i234/6395048 by DeepDyve user on 18 July 2022 i240 Annals of Work Exposures and Health, 2022, Vol. 66, No. S1 Figure 1. Modeled dermal exposure estimates of oil components (total hydrocarbons, benzene, ethylbenzene, toluene, xylene, n-hexane, and polycyclic aromatic hydrocarbons by broad activity groups by time period). Note that the scales (in GDUs) vary among the substances. ROVs = vessels piloting remotely operated vehicles. GDU = GuLF DREAM unit. (AM  = 2.34 GDU (2.00, 2.73 GDU); AM  = 1.28 GDU); AM  = 11.64 GDU (11.27, 12.02 GDU); and TP1a TP1b TP5 GDU (1.25, 1.31 GDU); AM  = 1.09 GDU (1.06, 1.12 AM  = 14.37 GDU (13.68, 15.09 GDU)). TP2 TP6 GDU); AM  = 1.09 GDU (1.08, 1.11 GDU)), but ex- In a comparison of the AMs across the broad groups, TP3 posures increased in TP4 and rose substantially in the land workers generally had statistically higher expos- last 2 time periods (AM   =  1.17 GDU (1.15, 1.19 ures than did other water workers, who generally had TP4 Downloaded from https://academic.oup.com/annweh/article/66/Supplement_1/i234/6395048 by DeepDyve user on 18 July 2022 Annals of Work Exposures and Health, 2022, Vol. 66, No. S1 i241 Figure 2. Modeled dermal exposure estimates of tar components (total hydrocarbons, benzene, ethylbenzene, toluene, xylene, n-hexane, and polycyclic aromatic hydrocarbons by broad activity groups by time period). Note that the scales (in GDUs) vary among the substances. ROVs = vessels piloting remotely operated vehicles. GDU = GuLF DREAM unit. statistically higher exposures than the rig, ROV, burner fire significant, the differences in the AMs for these workers control and RV workers, whether for oil or for tar (Fig. 1 were low and may not be meaningful. and 2 and SM, Table S6). Although often statistically Downloaded from https://academic.oup.com/annweh/article/66/Supplement_1/i234/6395048 by DeepDyve user on 18 July 2022 i242 Annals of Work Exposures and Health, 2022, Vol. 66, No. S1 For BTEX-H, the activities generally associated with PAHs, in contrast, had the highest maximums of ~219 the minimum values (<0.02 GDUs for all five chemicals) and ~588 GDUs for oil and tar, respectively. The broad varied (SM, Table S6). Benzene AMs for oil and for tar ranges of estimates for the exposures suggest that there across study participants reached 12.77 and 3.69 GDUs, should be sufficient contrast to differentiate between low respectively. AMs for ethylbenzene rose to 12.17 GDUs and high exposed individuals should these levels be asso- for oil and to 11.65 GDUs for tar. Similarly, the max- ciated with a health outcome. imum of the toluene AMs from oil was 17.45 GDUs; for There were statistically significant differences in the tar, the values rose to 42.37 GDUs. The maximums for AM exposures to THC in oil over time for the other xylene were 36.77 GDUs for oil and 88.18 GDUs for tar. water workers after TP1b and for land workers after The maximum AMs for n-hexane in oil were 2.22 GDUs TP2. Similarly, for tar, other water workers’ THC AMs and in tar, 5.56 GDUs. For PAHs, the maximum values rose after TP3, and land worker’s THC AMs rose after were 219.31 GDUs for oil and 587.98 GDUs for tar. TP4. We also evaluated THC exposure differences over Workers on the rig vessels, the ROV vessels and the broad groups of workers. Land workers had statistically burner fire control vessels were not considered exposed higher THC AM exposures from oil than did all other to dispersants (Table 2). The two components evaluated workers in TP1a to TP4. A similar pattern was seen for for dispersants were THC and xylene to represent pet- tar: land workers had higher THC AMs from tar in TP3, roleum distillates, hydrotreated light. The AM of the TP4, TP5, and TP6 than other water workers. The fact estimates for participants on land for THC were 25.41 that we found some differences suggests that we were and 18.88 GDUs and for xylene, 0.21 and 0.36 GDUs able to, at least to some degree, increase the accuracy of in TP1a and TP1b, respectively, the only time periods our estimates by considering jobs/activities/tasks, loca- for which dispersants were used. For participants on tion and time, thereby reducing misclassification error the water the AMs were for THC, 80.37 and for xylene, within the study participants. 1.39 GDUs, respectively. It seems counterintuitive that exposures to oil and Correlations among the substances in oil across all (for land workers) to tar should have risen over time. time periods combined were generally high (Table 1). The model is complex, however, and has many variables All substances had correlations of ρ ≥ 0.9 with each that affect the estimates, making it difficult to identify other except n-hexane. The n-hexane correlation with the single reason for this finding. In the online SM, we both THC and toluene was ρ = 0.87, with ethylbenzene, explore the effect of concentration, vapor pressure and rho = 0.84, xylene, ρ = 0.83 and with PAHs, ρ = 0.82. viscosity to learn how these varied by substance from Correlations in tar remained high for (ρ > 0.9) for the 25% weathering (around TP2, generally reflecting lower relationships between toluene, xylene, n-hexane and exposures from oil) to 40% weathering (TP5 and 6, re- PAHs. The relationships of THC and benzene, however, flecting higher exposures to tar). Viscosity appeared to with each other and with the other substances of interest be the predominant variable for ethylbenzene, toluene, were lower (ρ = 0.28–0.86 for THC, ρ = 0.70–0.89 for xylene and PAHs. Viscosity results in stickiness and the benzene). stickier the substance, the longer it will stay on the skin, thus increasing exposure. VP was the most important variable for THC and n-hexane. For both substances, Discussion the VP decreased substantially as the oil became more We describe procedures for the development of dermal weathered, resulting in an increase in the VP score (the exposure estimates derived from an updated and en- score being the inverse of VP) seen only with THC, ben- hanced version of DREAM (Gorman Ng et al., 2021). zene, and n-hexane, and therefore an increase in the We used participants’ interview responses when avail- exposure for THC and n-hexane. This decrease in VP able and JEM or industrial hygienist-entered values and the increase in viscosity should have resulted in an when not available. Even after deleting values <0.01 increase in benzene exposures over time; however, con- GDUs, minimum values were <0.05 GDUs for all sub- centration dominated the change seen in benzene by stances in oil or tar. For THC, values ranged up to ~5 being the second largest decrease among the substances. GDUs for oil and ~142 GDUs for tar. n-Hexane had See Stenzel, Arnold et  al. (2021) for more details on GDU levels of the same order of magnitude as those of weathering and its impact on the oil components. Less THC from oil, whereas the estimates for benzene, ethyl- of a contrast between benzene, THC, and n-hexane benzene, toluene, and xylene were substantially higher may have been seen had the differences in the VP score than THC for oil (up to 36.77 GDUs for xylene). For categories more closely reflected the differences in the VP tar, exposures were lower for the BTEX-H chemicals. (i.e. for 0–40% weathering, THC VP changed from 3986 Downloaded from https://academic.oup.com/annweh/article/66/Supplement_1/i234/6395048 by DeepDyve user on 18 July 2022 Annals of Work Exposures and Health, 2022, Vol. 66, No. S1 i243 Table 1. Correlations (ρ) between total hydrocarbons, benzene, ethylbenzene, toluene, xylene, n-hexane and polycyclic aromatic hydrocarbons in oil and tar. Correlation (ρ) among oil-related components (N = 1643 ) THC Benzene Ethylbenzene Toluene Xylene n-Hexane PAHs THC 1.00 0.98 >0.99 >0.99 0.98 0.87 0.98 Benzene 0.98 1.00 0.97 0.97 0.95 0.93 0.95 Ethylbenzene 1.00 >0.99 0.98 0.84 0.98 Toluene 1.00 0.98 0.87 0.98 Xylene 1.00 0.83 >0.99 n-Hexane 1.00 0.82 PAHs 1.00 Correlation (ρ) among tar-related components (N = 773) THC Benzene Ethylbenzene Toluene Xylene n-Hexane PAHs THC 1.00 0.28 0.77 0.68 0.77 0.86 0.80 Benzene 1.00 0.83 0.89 0.82 0.70 0.79 Ethylbenzene 1.00 >0.99 >0.99 0.97 >0.99 Toluene 1.00 >0.99 0.94 0.98 Xylene 1.00 0.96 >0.99 n-Hexane 1.00 0.97 PAHs 1.00 THC: total hydrocarbons. PAHs: polycyclic aromatic hydrocarbons. N = number of exposure groups. Table 2. Estimates of dispersant exposures (in GDUs). Chemical Broad group TP1a TP1b AM (LCL, UCL) AM (LCL, UCL) THC All land 25.41 (22.32, 28.93) 18.88 (17.26, 20.64) THC All other water operations NA 80.37 (62.54, 103.30) Xylene All land 0.21 (0.19, 0.24) 0.36 (0.33, 0.38) Xylene All other water operations NA 1.39 (1.08, 1.79) GDUs: GuLF DREAM units (see text for definition.) TP1a = time period 1a (April 22–May 14, 2010). TP1b = time period 1b (May 15–July 15, 2010). AM (AMLC, AMUC) = arithmetic mean (AM lower confidence value, AM upper confidence value). THC = total hydrocarbons. Workers on rig vessels, vessels piloting remotely operated vehicles (ROVs), burner fire control vessels, and research vessels were not identified as having dispersant exposures. NA = not applicable. No water workers in TP1a reported having contact with dispersants. to 821 Pa, respectively, whereas the scores changed from TP2 and in TP5 versus the median for activities only 0.01 to 0.1, respectively. For n-hexane, the respective performed in TP2 and found a higher median THC ex- values were 1025 to 36 Pa, with respective scores of 0.01 posure in the former (i.e. the first comparison of TP2 and to 1.0. For benzene, in contrast, those same respective TP5). In addition, the later activities were likely more as- values were 115 to 29 Pa but scores of 0.1 to 1.) sociated with tar (with the higher viscosity) than with oil A second reason for the rise in exposures is likely an (with the lower viscosity). artifact of the data, in that during the later time periods, The information used as inputs to the model was higher exposed activities comprise a larger percentage primarily from self-reports of the study participants for of the activities being performed. We compared the TP2 several reasons. We were able to observe only a limited median THC exposures of the activities performed in number of operations, as almost all OSRC activities Downloaded from https://academic.oup.com/annweh/article/66/Supplement_1/i234/6395048 by DeepDyve user on 18 July 2022 i244 Annals of Work Exposures and Health, 2022, Vol. 66, No. S1 had ceased by the time the GuLF STUDY was initiated. in the past. It is likely that reporting of activities in our In addition, there were thousands of workers across 4 study would be easier to recall than job classifications, states, and the oil was detected over more than 112 100 particularly with the short time interval between the spill 2 2 km (69  656 mi ) of water (Westerholm and Rauch, and the enrollment interview 1–3 years later and the un- 2016). Moreover, although there was guidance on what usual circumstances of the event. protective equipment to wear for some of the activities, We did not ask about changes over time. We know ac- on-site industrial hygienists were responsible for modi- tivities changed, as did the weathering of oil. While we fying the recommended equipment given the specific made allowances for changes in weathering, we do not work conditions encountered on any given day. know if the specific tasks performed for a particular ac- The estimates for the various oil-related substances tivity changed over time and had no way to adjust for generally were highly correlated (ρ > 0.9) for oil. In con- such an occurrence if it had transpired. It is unlikely how- trast, for tar the relationships between THC and ben- ever, that the respondents would have been able to differ- zene found lower correlations (ρ = 0.3–0.9), although entiate among exposure conditions in each of up to the the relationships among the other substances remained seven time periods of the study. Furthermore, participants high. The high correlations in general are not surprising, most often completed the enrollment questionnaire by however, because the variables associated with the re- telephone and the investigators were reluctant to add such ported job/activity/task were assigned the same weights detail to an already long interview. To reduce the potential regardless of the substance. The differences only came misclassification associated with not obtaining participant with substance-related variables, i.e. the concentration, information for the relevant time periods, we imputed vapor pressure score, and viscosity score. This finding missing data based on the responses from only those indi- may make it difficult to identify if a particular substance viduals who worked in each specific time period. This ap- is uniquely associated with an adverse health effect. proach could have biased estimates in other time periods It is not known how accurate the participants’ re- if participants tended to report the characteristics of the ports were. First, we asked about skin/clothing contact period with the greatest, or the least, skin burden. with tar, defined in the interview as a “solid or gooey Estimates for some of the variables were assigned oily residue”, and then with oil (“oil or oily water”) to by the study industrial hygienists. Of these, some were reduce the likelihood of participants confusing oil with substance-specific data (concentration, vapor pressure tar. Some participants still, however, may have confused score, viscosity score) that changed with the degree of the two, particularly after the oil had undergone some weathering. Other variables were the percentages of the weathering. This confusion primarily would have af- body parts exposed and the emission pathway, both of fected participants who worked on land, although par- which we did not think that respondents would have been ticipants performing water activities after TP3 (August able to provide. We had no way to validate our assign- 10–September 1, 2010) could also have been confused. ments because the oil spill response and clean-up effort Depending on the substance, this confusion could result was largely completed by the time the exposure assess- in an over- or underestimation of exposures. Second, we ment started, and we were able to observe few workers asked about frequency of skin exposure in terms of the during the performance of their jobs. We reviewed, proportion of a day (none, less than half, about half, however, the extensive amount of documentation, par- more than half, all of it), but we asked about hand ex- ticularly on weathering (see the SM in Stenzel, Arnold posure in terms of hours and minutes. In the home visit, et al., 2021) and the substantial number of photographs we asked how often contact with “a chemical” occurred taken of OSRC workers to obtain information on work with each body part (<1 day per month, 1-4 days per activities. month, 1–5 days per week or almost every day), and we We did not evaluate the day-to-day variability of ex- assumed that the frequency on the body part would have posures experienced by an individual due to the lack of been the same for all exposures of interest, which likely day-to-day information. The model developed point esti- introduced some error in the exposure estimation. It is mates from over 50 exposure variables. Given the amount not clear what the magnitude would be and whether the of uncertainty in the self-reports, this additional uncer- error would result primarily in an over or underestima- tainty was unlikely to have provided useful information. tion of exposure. In contrast, reports of jobs/activities/ In our evaluation of the GuLF DREAM model, we tasks was likely to have been good. Teschke et al. (2002), were unable to validate our actual estimates with meas- found that reporting of job classifications generally had urements taken during the OSRC, and we were unable good agreement with records (Teschke et al., 2002), and to find any useful dermal measurements taken on oil recollection was better for recent jobs than jobs further spill workers. The only oil spill dermal measurements we Downloaded from https://academic.oup.com/annweh/article/66/Supplement_1/i234/6395048 by DeepDyve user on 18 July 2022 Annals of Work Exposures and Health, 2022, Vol. 66, No. S1 i245 found were from a National Institute for Occupational and high-exposed workers to distinguish between the Safety and Health (NIOSH) study of workers responding most heavily and least exposed type of workers. to the Exxon Valdez spill (Gorman et al., 1991), which reported that dermal exposure levels were higher prior Conclusions to the work shift than after the work shift. Thus, we used two published datasets: one, for oil, represented by Assessment of dermal exposures is difficult, and in this a study measuring heavy fuel oil exposures in a variety study, there were no dermal measurements made during of oil-using industries (Christopher et al. 2007, 2011) the response and clean-up operations. We used a deter- and the other, for tar, represented by a study of an as- ministic model with participant-reported and industrial phalt paving operation (Cavallari et al., 2012). Overall, hygienist-supplied data. Dermal exposures to seven sub- the evaluation demonstrated a moderate correlation stances contained in oil and tar and to dispersants were (ρ = 0.59) between GuLF DREAM exposure estimates estimated for each of the jobs/activities/tasks by location and hand wash and wipe measurements. This correlation and time period reported by the nearly 24 000 workers in was somewhat lower than the corresponding correlation the study. The range of dermal estimates was substantial calculated for DREAM (ρ = 0.78) (van Wendel de Joode, for many of the exposures of interest. This provides some 2005). This is not surprising because the DREAM valid- confidence that there may be enough contrast to identify ation by van Wendel de Joode (2005) involved exposure exposure-related differences in workers’ health, should assessors who had often directly observed the workers they exist; however, the high correlation among most of during the exposure measurements. GuLF DREAM was the exposures makes it unclear whether we will be able to evaluated with previously collected datasets of other in- identify a particular substance among the many assessed. vestigators, so that study investigators did not have the Dermal exposure assessment is still in its infancy and the opportunity to observe the measured workers directly. usefulness of deterministic models needs further evaluation. We were unable to evaluate the model for body parts other than hands due to insufficient measurements in the Supplementary data published studies. The original DREAM model found poorer correlation with other body parts. Supplementary data are available at Annals of Work Exposures Limitations of this work include dependence on the and Health online. study participants’ self-reports; our inability to observe the response and clean-up activities; lack of time-varying information on specific work; lack of information on Acknowledgments variability; and lack of an evaluation on parts of the We thank Wendy McDowell and Kaitlyn Rousch of McDowell body other than hands. In addition, the reports of dis- Safety and Health Services, Inc. and Matthew Curry, Braxton persants may have been overreported; positive responses Jackson, John McGrath and Kate Christenbury of Social & Scientific suggest more participants had exposure than expected. Systems, Inc. for the tremendous help they provided on this study. Finally, we estimated exposure using a dimensionless We also thank the workers for their participation in this study. unit and do not know how it relates to absorption. This is, however, the first study to provide relative Funding dermal exposure estimates from working on the response and clean-up of an oil spill under the many varied activ- This study was funded by the NIH Common Fund and the ities that took place during the OSRC. We used a deter- Intramural Research Program of the National Institute of Health, National Institute of Environmental Health Sciences ministic model, an approach that has been used in other (ZO1 ES 102945). studies without measurement data (Vermeulen et  al., 2002). The evaluation, although limited, provides some information on both oil and asphalt (similar to tar) ex- Conflict of interest posures. The dermal estimates show a different exposure Prof Cherrie is currently undertaking consulting work related pattern from inhalation, allowing investigators to inves- to the Deepwater Horizon disaster. All of his involvement with tigate different routes of exposure and disease mechan- this paper was prior to any potential conflict of interest arising. isms. Finally, the estimates yielded large contrasts across the range of exposure levels for the substances of interest. This broad range, while likely to contain misclassification Data availability error, should allow investigators to examine risks associ- The data underlying this article will be shared on reasonable ated with categories of exposure, such as low-, medium-, request, consistent with protections for the privacy of study Downloaded from https://academic.oup.com/annweh/article/66/Supplement_1/i234/6395048 by DeepDyve user on 18 July 2022 i246 Annals of Work Exposures and Health, 2022, Vol. 66, No. S1 participants and existing multi-party agreements. Requests Deepwater Horizon oil spill clean-up operations. Ann Work should be made following instructions on the study website Expo Health; 66: i124–i139. https://gulfstudy.nih.gov. Kwok RK, Engel LS, Miller AK et al.; GuLF STUDY Research Team. (2017) The GuLF STUDY: a prospective study of per- sons involved in the deepwater horizon oil spill response References and clean-up. Environ Health Perspect; 125: 570–8. Arnold S, Stewart PA, Pratt GC et al. (2021) Estimation of aerosol NIOSH. (2011) NIOSH Deepwater Horizon Roster Summary TM Report. Available at https://www.cdc.gov/niosh/updates/ concentrations of oil dispersants COREXIT EC9527a and EC9500A during the Deepwater Horizon oil spill re- upd-12-19-11.html. (accessed 16 June 2018). NIOSH. (1994) Pocket Guide to Chemical Hazards. DHHS sponse and clean-up operations. Ann Work Expo Health; 66: i188–i201. (NIOSH) Publication No. 94-116. Pratt GC, Stenzel MR, Kwok RK et al. (2021) Modeled air pol- Cavallari JM, Osborn LV, Snawder JE et al. (2012) Predictors of dermal exposures to polycyclic aromatic compounds lution from in situ burning and flaring of oil and gas re- leased following the Deepwater Horizon disaster. Ann Work among hot-mix asphalt paving workers. Ann Occup Hyg; 56: 125–37. Expo Health; 66: i172–i187. Ramachandran G, Groth CP, Huynh TB et  al. (2021) Using Christopher  Y, van  Tongeren  M, Cowie  H et  al. (2007) Occupational dermal exposure to heavy fuel oils. Research real-time area VOC measurements to estimate total hydro- carbons exposures to workers involved in the Deepwater Report TM/07/05. Edinburgh, UK: IOM. Christopher Y, Van Tongeren M, Urbanus J et al. (2011) An as- Horizon oil spill. Ann Work Expo Health; 66: i156–i171. Stenzel MR, Arnold SF, Ramachandran G et al. (2021) Estimation sessment of dermal exposure to heavy fuel oil (HFO) in oc- TM cupational settings. Ann Occup Hyg; 55: 319–28. of airborne concentrations of oil dispersants COREXIT EC9527A and EC9500A, volatile components associated Gorman RW, Berardinelli SP, Bender TR. (1991) Exxon Valdez Alaska Oil Spill (1991) HETA 89-200 & 89-273-2111. with the Deepwater Horizon oil spill response and clean-up operations. Ann Work Expo Health; 66: i202–i217. Cincinnati, OH: National Institute for Occupational Safety and Health. Stenzel MR, Groth CP, Huynh TB et al. (2021) Exposure group development in support of the NIEHS GuLF STUDY. Ann Gorman  Ng  M, Cherrie  JW, Sleeuwenhoek  A et  al. (2021). GuLF DREAM: a model to estimate dermal exposure Work Expo Health; 66: i23–i55. Stewart P, Groth C, Huynh TB et al. (2021) Assessing expos- among oil spill response and clean-up workers. Ann Work Expo Health; 66: i218–i233. ures from the Deepwater Horizon oil spill response and clean-up. Ann Work Expo Health; 66; i3–i22. Groth CP, Banerjee S, Ramachandran G et al. (2021) Methods for the analysis of 26 million VOC area measurements Teschke K, Olshan AF, Daniels JL et al. (2002) Occupational ex- posure assessment in case–control studies: opportunities for during the Deepwater Horizon oil spill clean-up. Ann Work Expo Health; 66: i140–i155. improvement. Occup Environ Med; 59: 575–594. Van Wendel de Joode B, Brouwer DH, Vermeulen R et al. (2003) Groth CP, Huynh TB, Banerjee S et al. (2021) Linear relationships between total hydrocarbons and benzene, toluene, ethylben- DREAM: a method for semi-quantitative dermal exposure assessment. Ann Occup Hyg; 47: 71–87. zene, xylene, and n-hexane during the Deepwater Horizon response and clean-up. Ann Work Expo Health; 66: i71–i88. Van Wendel de Joode B, Vermeulen R, van Hemmen JJ et al. (2005) Accuracy of a semiquantitative method for Dermal Exposure Huynh TB, Groth CP, Ramachandran G et al. (2021a) Estimates of occupational inhalation exposures to six oil-related com- Assessment (DREAM). Occup Environ Med; 62: 623–32. Vermeulen R, Stewart P, Kromhout H. (2002) Dermal exposure pounds on the four rig vessels responding to the Deepwater Horizon oil spill. Ann Work Expo Health; 66: i89–i110. assessment in occupational epidemiologic research. Scand J Work Environ Health; 28: 371–85. Huynh TB, Groth CP, Ramachandran G et al. (2021b) Estimates of inhalation exposures to oil-related components on the Westerholm  DA, Rauch  SD III. (2016) Deepwater Horizon oil spill: final programmatic damage assessment and res- supporting vessels during the Deepwater Horizon oil spill. Ann Work Expo Health; 66: i111–i123. toration plan and final programmatic environmental im- pact statement. https://repository.library.noaa.gov/view/ Huynh TB, Groth CP, Ramachandran G et al. (2021c) Estimates of inhalation exposures among land workers during the noaa/18084 (accessed 6 June 2020) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Annals of Work Exposures and Health (formerly Annals Of Occupational Hygiene) Oxford University Press

Estimation of Dermal Exposure to Oil Spill Response and Clean-up Workers after the Deepwater Horizon Disaster

Loading next page...
 
/lp/oxford-university-press/estimation-of-dermal-exposure-to-oil-spill-response-and-clean-up-WTPgZXfBGp

References (19)

Publisher
Oxford University Press
Copyright
Copyright © 2022 British Occupational Hygiene Society
ISSN
2398-7308
eISSN
2398-7316
DOI
10.1093/annweh/wxab073
Publisher site
See Article on Publisher Site

Abstract

The GuLF STUDY is investigating health outcomes associated with oil spill-related chemical ex- posures among workers involved in the spill response and clean-up following the Deepwater Horizon disaster. Due to the lack of dermal exposure measurements, we estimated dermal ex- posures using a deterministic model, which we customized from a previously published model. Workers provided information on the frequency of contact with oil, tar, chemical dispersants ap- plied to the oil spill and sea water, as well as the use of protective equipment, by job/activity/ task. Professional judgment by industrial hygienists served as a source of information for other model variables. The model estimated dermal exposures to total hydrocarbons (THC), benzene, ethylbenzene, toluene, xylene, n-hexane (BTEX-H), polycyclic aromatic hydrocarbons (PAHs), and dispersants in GuLF DREAM units (GDUs). Arithmetic means (AMs) of THC exposure estimates across study participants ranged from <0.02 to 5.50 GDUs for oil and <0.02 to 142.14 GDUs for tar. Statistical differences in the estimates were observed among the AMs of the estimates for some broad groups of worker activities over time and for some time periods across the broad Published by Oxford University Press on behalf of The British Occupational Hygiene Society 2021. Downloaded from https://academic.oup.com/annweh/article/66/Supplement_1/i234/6395048 by DeepDyve user on 18 July 2022 Annals of Work Exposures and Health, 2022, Vol. 66, No. S1 i235 groups of activities. N-Hexane had ranges similar to THC for oil exposures (e.g. AMs up to 2.22 GDUs) but not for tar (up to 5.56 GDUs). Benzene, ethylbenzene, toluene, and xylene, in contrast, were characterized by higher exposure levels than THC for oil (AMs up to 12.77, 12.17, 17.45, and 36.77 GDUs, respectively) but lower levels than THC to tar (AMs up to 3.69, 11.65, 42.37, and 88.18 GDUs, respectively). For PAHs, the AMs were as high as 219.31 and 587.98 for oil and tar, re- spectively. Correlations of these seven substances to each other were high (>0.9) for most of the substances in oil but were lower for some of the substances in tar. These data were linked to the study participants to allow investigation of adverse health effects that may be related to dermal exposures. Keywords: Deepwater Horizon; dermal exposure, exposure assessment; oil spill; total hydrocarbons Introduction This paper describes the methods and the results for the dermal assessments. On 20 April 2010, the Deepwater Horizon oil rig ex- An overview of the exposure assessment effort ploded in the Gulf of Mexico, causing almost 5 million for the STUDY is presented in Stewart et  al. (2021). barrels of oil to be released into the Gulf waters over the Development of exposure groups (EGs) is described in following 3 months. Over 55 000 workers were rostered Stenzel, Arnold et al., 2021. The assessment of airborne by NIOSH as having participated in the response and exposures to THC and BTEX-H is described in Huynh clean-up (NIOSH, 2011). Workers had inhalation and et al., 2021a,b,c; Ramachandran et  al., 2021; Groth, dermal exposures to multiple oil-related compounds, as Banerjee et al., 2021; and Groth, Huynh et al., 2021. well as possible exposure to chemical dispersants, PM , 2.5 Assessment of other airborne exposures is also reported and cleaning products. Although more than 160 000 air (PM (Pratt et al., 2021); dispersant aerosols (Arnold 2.5 measurements were available to characterize inhalation et al., 2021) and vapors (Stenzel, Groth et al., 2021); and exposures, no dermal or surface wipe measurements oil mists (Stewart et al., 2021)). had been collected. Furthermore, few measurements of dermal exposure were available from other spills for ex- posure characterization. Background The Gulf Long-term Follow-up Study (GuLF The Deepwater Horizon (DWH) oil spill led to a mas- STUDY), initiated by the National Institute of sive effort to contain the spill and clean the Gulf of Environmental Health Sciences (NIEHS), is Mexico waters and shoreline. Most of the OSRC ac- investigating potential adverse health effects associated tivities were suspected as having resulted in dermal with the oil spill response and clean-up (OSRC) (Kwok exposure to oil, oily salt water, and tar. Two rigs (the et  al., 2017). As part of the exposure assessment ef- Discoverer Enterprise (Enterprise) and the Helix fort, we updated and enhanced a previously published Q4000 (Q4000)), were involved in mitigating the re- dermal exposure deterministic model (Van Wendel de lease, capturing the leaking oil/natural gas mixture; Joode et al., 2003) to better reflect the contribution of and separating the gas from the oil and flaring the gas various exposure determinants relevant to the GuLF (Enterprise) or the oil/gas mixture (Q4000) Huynh et al., STUDY (Gorman Ng et al., 2021). Using information 2021a. These rigs were located within 1 nautical mile from both the study participants and the study indus- (nmi, 1.8 km; 1.1 mi) of the wellhead, approximately trial hygienists, we estimated exposure to total hydro- 50 nmi (93 km) southeast of the Louisiana (LA) shore. carbons (THC), benzene, toluene, ethylbenzene, xylene, Two other drilling rigs, the Development Driller II n-hexane (BTEX-H), polycyclic aromatic hydrocarbons (DDII) and the Development Driller III (DDIII), lo- (PAHs) (as a single substance), and (total) dispersants cated within the 1 nmi radius of the wellhead (referred due to these substances’ inhalation toxicity and their to as the hot zone), were each responsible for drilling a ability to be absorbed into the skin or adversely affect relief well. the skin. (Dispersant refers to chemicals sprayed onto Supporting these four vessels was a sizable, but un- an oil slick on the water surface or injected into the known, number of other large marine vessels (MVs) water to break down the oil into small droplets that (Huynh et al., 2021c; Ramachandran et  al., 2021). more readily mix with the water. It is the dispersants’ Fourteen MVs that piloted remotely operated vehicles components that are associated with possible toxicity.) Downloaded from https://academic.oup.com/annweh/article/66/Supplement_1/i234/6395048 by DeepDyve user on 18 July 2022 i236 Annals of Work Exposures and Health, 2022, Vol. 66, No. S1 (ROVs), called here ROV vessels, performed several Methods underwater activities, such as moving equipment, col- After review of several dermal assessment models, the lecting water samples and taking videos. Other MVs DREAM model (Van Wendel de Joode B et al., 2003) provided other types of support, for example, pumping was selected as the most appropriate model for the fluids into the well for well closure attempts; spraying GuLF STUDY because it allowed the use of information water onto the flaring vessels to reduce temperatures; on determinants of dermal exposure, some of which storing and transporting collected oil; supplying ma- had been collected from study participants through a terials/chemicals/crew; and spraying dispersant onto telephone interview. Validation work conducted by the the water’s surface near the rigs. We called the 5 nmi (9 developers of the original model found that the dermal km) radius around the wellhead, excluding the hot zone, exposure units (DREAM units, a dimensionless unit) where most of the supporting vessels worked, the source. correlated well with hand exposure measurements Research vessels collected samples of water and across a range of work sites and exposure agents, but oil, took water quality measurements, monitored the less well with other body parts (Van Wendel de Joode B oil plume and collected oiled and dead wildlife. Other, et al., 2005). typically smaller, vessels, such as fishing and shrimping We updated the DREAM model to reflect more recent boats, scouted for oil; deployed/maintained/retrieved studies and adapted it, primarily by changing weights of booms; skimmed or burned oil off the water surface; the various determinants to fit the oil-related chemicals collected wildlife; carried personnel, equipment and and dispersants and available data of the GuLF STUDY supplies to and from the wellhead area; and transported to allow retrospective evaluation using interview data collected oil and oily water back to shore Huynh et al., from study participants (Gorman Ng et al., 2021). We 2021b. Vessels also decontaminated (deconned) other called this model GuLF DREAM and the dimensionless vessels, jetties and other manmade structures using pres- outcome measures, GuLF DREAM units (GDUs). The sure spraying. Operations occurred throughout the Gulf model required variables for: of Mexico, north of the wellhead off the LA, Mississippi (MS), Alabama (AL), and Florida (FL) coastlines. • the concentration, vapor pressure score, and viscosity Activities on land (Huynh et al., 2021c) included score of each substance; loading/unloading of everything related to the vessels’ • each of three exposure pathways: emission (direct functions described above at ports and docks, primarily contact with the liquid substance), deposition (con- in LA, MS, AL, and FL. Deconning of vessels, equipment tact with the airborne substance), and surface on the vessels, boom and other gear was done by low- or transfer (contact with a contaminated surface); high-pressure spraying or by hand (e.g. rags or absorb- • the surface area of each of nine body parts (head, ents). Beaches and marshes were systematically patrolled upper arm, lower arm, hands, front of torso, back of to evaluate locations needing attention, after which they torso, upper legs, lower legs, and feet); were cleaned by picking up oil; oily water; tar balls, • for each body part, the intensity (i.e. the percentage patties, and mousse; oiled plants; and garbage by hand or of surface area covered) and the frequency of ex- using handheld equipment for disposal in bags. Wildlife posure for the likeliest pathway for oil and tar; was captured and rehabilitated. Support staff included ma- • water (seawater removes water soluble compounds) terial handling, cooks, housekeeping, office, and security contact; and workers. • the type and frequency of replacement of gloves and As the oil was released from the well, it changed com- protective clothing. position, due to natural processes including evaporation, Each variable was assigned a value, generally between 0 dispersion, emulsification, dissolution, photo-oxidation, and 5 (Gorman Ng et al., 2021). sedimentation, and biodegradation. These processes oc- To evaluate our model, exposures measured in two curred from the time the oil was released in the subsea studies of heavy fuel oil (in a variety of industries unre- water and continued while it was on the water surface lated to oil spills (Christopher et al., 2007)) and of as- and on land. This change in oil, called weathering, re- phalt (among paving workers (Cavallari et al., 2012)) sulted over time in a differential decrease of the percent- were estimated using GuLF DREAM and compared to ages of the volatile chemicals and a differential increase those studies’ measurements. A correlation (ρ) of 0.59 in the percentages of the semi- and non-volatile chem- was found between the GuLF DREAM estimates and the icals in the oils and tars (Stenzel, Arnold et al., 2021). measurements for the hands (Gorman Ng et al., 2021). Downloaded from https://academic.oup.com/annweh/article/66/Supplement_1/i234/6395048 by DeepDyve user on 18 July 2022 Annals of Work Exposures and Health, 2022, Vol. 66, No. S1 i237 There were insufficient measurements to evaluate other Questions on the participant’s use of protective equip- body parts. ment by job/activity/task covered the use of: Because dermal exposure was likely to have occurred • leather, cotton, or synthetic gloves; and if a second both to oil and to tar, we assessed a variety of chemicals glove was worn; contained in each of those two substances, i.e. THC, as • boots or rubber slip-ons; petroleum hydrocarbons; each of the BTEX-H chemicals; • protective coveralls such as Tyvek; and PAHs as a single substance. For dispersants, based on • long sleeved shirts, jackets, or coveralls; and the relative quantity of each dispersant used, we evaluated • the frequency of changing each of these types of a 33.4/66.6 mixture of COREXIT™ 9527A and 9500A clothing. (based on the amounts of the two dispersants used) applied by air across the Gulf and 9500A alone at the wellhead, ei- All questions were asked of all study participants, except ther sprayed on the water or injected near the seabed. for the set of questions on the frequency of exposure to an unspecified chemical for the eight body parts, which was Data collection only asked of the home interview participants. See the on- In the GuLF STUDY telephone interview, participants line Supplemental Material (SM), Table S1 for more details. (N = 24 937) were asked about the variety of jobs/activ- Processing of information for the estimates ities/tasks presented in Background, as well as questions that addressed dermal exposure specific to most of the Data coding A set of rules was developed to translate the question re- reported job/activities/tasks (for exceptions, see below). A subset of the participants (N = 11 193) later com- sponses into weights for the GuLF DREAM model (Gorman Ng et al., 2021). The same weights were assigned for oil, pleted a home visit interview that included other dermal questions. The dermal questions asked were: tar and dispersants. For example, if a participant said yes to “Did your skin or clothing come in contact with (substance) • for each activity reported by the participant, if his/ during any of your oil spill clean-up work?”, a value of 1 was her skin or clothing had contact with each of several assigned if the response was yes and 0 if the response was substances, including oil, tar, dispersants, and water. no. If yes, a follow-up question was asked, “On an average (We were unable to find a definition for the difference workday, how much of the time was your skin or clothing in between oil and tar that we thought the participants contact with (substance)?” Responses were coded on a scale would be able to distinguish. We therefore defined of 1–5: “None” = 1, “<1/2” = 2, “About ½” = 3, “>1/2” = 4, the two substances as “a solid or gooey oily residue and “All of it” = 5. Rules were developed based on the sub- or tar” (asked first) and “oil or oily water”). stance being assessed, the question number, and the variable. Questions were missing responses for approximately If contact occurred, additional questions were asked for 5–10% of participants. In addition, only approximately each reported job/activity/task: one-third of the cohort, i.e. those who had home visits, • the frequency of the contact (none; less than half the had responses on frequency of contact to the eight body time; about half; more than half; or all the time); parts. To impute information from both types of missing • whether the substance got on the hands (yes/no); data, a job-exposure matrix (JEM) approach was taken. • the number of hours a day it was on the hands before The basis for the JEM was the set of EGs, unique com- being washed off (h, min); binations of jobs/activities/tasks, location, and time • whether the substances got on the skin or clothing (Stenzel, Arnold et al., 2021), developed for inhalation other than the hands (yes/no); exposures. Time periods (TPs) are described in the SM, • frequency of contact with water (none; less than Table S2 and jobs/activities/tasks in the SM, Table S3. half the time; about half; more than half; or all the Briefly, the events in each time period were: time); and • the frequency that each of the eight other body parts • TP1a (22 April through 14 May 2010): oil flowed became wet with a (unspecified) chemical (<1 day from the damaged well. Drilling started on a relief per month; 1–4 days per month; 1–5 days per week; well. Water clean-up activities started. Oil reached the or almost every day), whether on clothing or on the LA shoreline, and beach cleanup started. Dispersant skin. application began. Downloaded from https://academic.oup.com/annweh/article/66/Supplement_1/i234/6395048 by DeepDyve user on 18 July 2022 i238 Annals of Work Exposures and Health, 2022, Vol. 66, No. S1 • TP1b (15 May through 15 July 2010): oil flow the assigned JEM value was 1 (yes), and 0 (no) if <50% continued. Drilling began on a second relief well. of participants answered “yes”). Median integer scores Dispersant operations continued. Water clean-up ac- were used for categorical or continuous responses (e.g. tivities continued. The well was successfully mechan- for the question, “On an average workday, how much of ically capped on July 15, which essentially stopped the time was your skin or clothing in contact with oil?”, the release of oil. Beach and wildlife clean-up was the responses were coded as: “None” = 1, “<1/2” = 2, being done in all four states. “About ½” = 3, “>1/2” = 4, and “All of it” = 5. The in- • TP2 (16 July through 10 August 2010): the well was teger of the median score across all respondents was as- “static killed” on 10 August. Water activities started signed as the JEM value. diminishing. Beach and jetty clean-up and decontam- To minimize participant burden, questions were not ination of vessels and equipment continued. asked about changes over time. The JEM values for a • TP3 (11 August through 30 September 2010): given job/activity/task/location, however, varied based large-scale final decontamination of the vessels and on the responses of the participants who worked in related equipment started. Water activities continued each time period. For example, workers who reported to lessen. Beach clean-up continued but started to cleaning oil pools in LA in TP1a (N = 59) had a median decline. emission frequency of 3 (“1/2 the time”), whereas for • TP4 (1 October through 31 December 2010): water those who cleaned oil pools in TP1b (N = 461), the me- efforts essentially were completed by the end of dian emission frequency was 1 (“never”), although many December. Beach and marsh clean-up continued by of the 59 in TP1a also worked in TP1b. When <5 partici- decreasing numbers of people. pants or <10% of the participants provided responses to • TP5 (1 January to 31 March 2011)  and TP6 (1 an EG, the reported summary value was overridden and April to 30 June 2011): beach clean-up continued by replaced by the summary value for “All states” for that decreasing numbers of people. Water operations were activity and time period. limited to the near shore transfer of equipment, sup- Some GuLF DREAM variables were not sought plies, collected materials and personnel. TP6 is distin- from the study participants because the variables re- guished from TP5 because of the warmer ambient air quired information the participants were unlikely to temperatures. know. Instead, the study industrial hygienists supplied the information based on the scientific literature, know- Most of the inhalation EGs were retained, although ledge of exposures, extensive published and unpublished we modified some to incorporate additional informa- documentation on the DWH event, and consensus of the tion obtained from the interviews by adding groups to study hygienists. This information included three prop- account for differences relevant to dermal exposures. erties of the substances distinguished by the degree of For example, we had an inhalation EG for “Handled/ weathering: i.e. the concentration of the various com- cleaned wildlife”, which was as precise as the air meas- ponents in the oil or tar; a vapor pressure score; and a urement data allowed. For the dermal assessment, viscosity score. The values for these scores and their der- however, we distinguished among the various wildlife ivation are described in SM, Table S4. We also identified handling/cleaning activities, i.e. “Handled oily wild- the percentage of oil weathering associated with each life”, “Cleaned wildlife”, “Used soaps to clean wildlife” EG (SM, Table S5). and “Retrieved dead wildlife”. New EGs were also de- We assigned other exposure determinants. First, we veloped for workers on the rig vessels, because the job- assigned the primary pathway of exposure (emission or based inhalation EGs had small numbers of respondents. surface transfer) based on knowledge and photographs Small sample sizes would have resulted in less precise of the job/activity/task. We considered that the contribu- values when using study participants’ responses to im- tion of deposition was very small in light of the airborne pute missing data, so we combined the various jobs to concentrations measured relative to the contribution form 7 broad “jobs”. Descriptions of all jobs/activities/ from emission or surface transfer (the average of the tasks are provided in SM, Table S3. THC air vapor concentrations across EGs (Huynh et al., The data used to complete the JEM cells were sum- 2021a,b,c); Ramachandran et al., 2021 was 0.8 ppm). maries of the responses to each dermal-related question Thus, deposition was not considered a pathway and de- for each EG. Percentages were used for imputing yes/ position frequency and intensity were assigned “0”. no questions (e.g. if ≥50% of participants answered Second, the study industrial hygienists entered values “yes” to “Did your skin or clothing come in contact for the intensity received on each body part for the ap- with oil during any of your oil spill clean-up work?”, propriate pathway, i.e. emission or surface transfer. Downloaded from https://academic.oup.com/annweh/article/66/Supplement_1/i234/6395048 by DeepDyve user on 18 July 2022 Annals of Work Exposures and Health, 2022, Vol. 66, No. S1 i239 The DREAM definition of intensity was the amount of under weathering conditions of 25–30%, we combined the body part exposed, and we retained that definition the oil and tar estimates for any participant’s job/activity/ for GuLF DREAM. Intensity values and weights were: task, location and time period associated with those de- <10% of body part  =  1; 10–50% of body part  =  3; grees of weathering. This procedure primarily affected the ≥50% of body part = 10. A single intensity weight across latter time periods (TP3–6). Statistical differences in the all time periods was assigned to each body part for each AMs were identified by non-overlapping 95% confidence job/activity/task/location. intervals (upper confidence level, UCL; lower confidence Third, some types of clothing and wind speed values level, LCL). Pearson correlations were calculated among were assigned by the industrial hygienists. We assumed the various components of oil and of tar. headgear was cotton hats. We assumed that workers in all jobs/activities/tasks had been required to wear heavy (leather, cotton, or synthetic) gloves, except for security, Results general environment/land, housekeeping, kitchen, and The range of THC dermal AM estimates from oil ex- office workers, all of whom were assumed to have worn posure was broad, ranging from AMs <0.02 GDUs light (latex) gloves. Frequency of hat replacement was (among several groups, for example, “IH/safety-water”, considered less than daily and clothes and shoes (rubber “All states”, TP1a) to 5.50 GDUs (“Deconned booms/ booties), daily. The frequency of glove replacement, how- land”, “LA”, TP3) (not shown). Equivalent tar AMs were ever, was provided by the study participant: “less than <0.02 GDUs (such as “Ran mechanical equipment/ports daily”, “daily” or “within a work shift”. Wind speed was & docks” All states, TP3) to 142.14 GDUs (“Retrieving coded as 1 (i.e. no effect on the substance), except for boom in shallow water”, MS, TP6). dispersant, which was assigned 0.75 (small effect). For THC, no statistical differences in oil exposures Finally, participants were not asked about dermal occurred across time periods among the workers on exposures for jobs/activities/tasks that were expected to the rigs (Fig. 1 (note that the scales differ by substance) have no or very little dermal exposure, such as “Cooks”, and SM, Table S6) The values for the variables asso- “Office workers”, and “Security”. Other workers were ciated with the ROV, burner fire control and research linked to an EG from a response to an open-ended ques- vessels were assigned by the industrial hygienists and tion (“What else did you do?”) and therefore did not get therefore there was little variability in the estimates. asked dermal questions (e.g. workers on ROV and re- Other water workers were characterized by increasingly search vessels). The study industrial hygienists entered a higher (and statistically significant) mean dermal esti- single value for each variable across all states and time mates over time: AM  = 0.39 GDU (LCL: 0.38, UCL: TP1b periods to allow prediction for these jobs/activities/tasks. 0.39 GDU); AM   =  0.49 GDU (0.48, 0.50 GDU); TP2 Once all participants had complete information for AM  = 0.56 GDU (0.56, 0.57 GDU); AM   =  0.85 TP3 TP4 each job/activity/task, either from the responses or the GDU (0.83, 0.88 GDU); and AM  = 1.31 GDU (0.95, TP5 imputed data, we applied the GuLF DREAM model 1.83 GDU). For land workers, the AMs significantly fell to develop exposure estimates for each participant’s from TP (AM = 0.97 GDU (0.92, 1.01 GDU)) to TP 1a 1b unique combination of job/activity/task, location and and TP (AM for both time periods = 1.59 GDU (0.58, time period. Because multiple jobs/activities/tasks were 0.60 GDU)) and then started rising, with statistically reported (median=6 per participant), multiple estimates significant higher AMs in each of the later time periods were assigned per time period. (AM  = 0.92 GDU (0.91, 0.93 GDU); AM  = 1.16 TP3 TP4 GDU (1.14, 1.18 GDU); AM  = 1.32 GDU (1.27, 1.38 TP5 Statistical analyses GDU); and AM  = 1.67 GDU (1.56, 1.78 GDU). TP6 For presentation purposes, the estimated arithmetic means Tar exposure was assessed only for “Other water op- (AMs) of the GDUs across all study participants were erations” and “Land”, as the other vessels (rigs, ROVs, calculated by job/activity/task, location, time period and burner fire control vessels and RVs) had left the area substance. In the AM calculation, we dropped all job/ac- by the time the oil had likely weathered to tar. Water tivity/task, location, time period combinations that were workers were exposed to higher statistically signifi- <0.01 GDUs. For the graphs, we used the broad group- cant differences over time (AM  = 0.37 GDU (0.35, TP3 ings of workers for “All rigs”, “All ROVs”, “All research 0.38 GDU); AM   =  0.86 GDU (0.83, 0.89 GDU); TP4 vessels”, “Burner fire control vessels”, “Other water”, and AM  = 1.13 GDU (1.05, 1.21 GDU); and AM  = 1.46 TP5 TP6 “Land” by time period for “All states” for oil and for tar. GDU (1.31, 1.62 GDU)) (Fig. 2 and SM, Table S6). Due to our inability to distinguish between oil and tar Land workers’ tar exposures fell from TP1a to TP3 Downloaded from https://academic.oup.com/annweh/article/66/Supplement_1/i234/6395048 by DeepDyve user on 18 July 2022 i240 Annals of Work Exposures and Health, 2022, Vol. 66, No. S1 Figure 1. Modeled dermal exposure estimates of oil components (total hydrocarbons, benzene, ethylbenzene, toluene, xylene, n-hexane, and polycyclic aromatic hydrocarbons by broad activity groups by time period). Note that the scales (in GDUs) vary among the substances. ROVs = vessels piloting remotely operated vehicles. GDU = GuLF DREAM unit. (AM  = 2.34 GDU (2.00, 2.73 GDU); AM  = 1.28 GDU); AM  = 11.64 GDU (11.27, 12.02 GDU); and TP1a TP1b TP5 GDU (1.25, 1.31 GDU); AM  = 1.09 GDU (1.06, 1.12 AM  = 14.37 GDU (13.68, 15.09 GDU)). TP2 TP6 GDU); AM  = 1.09 GDU (1.08, 1.11 GDU)), but ex- In a comparison of the AMs across the broad groups, TP3 posures increased in TP4 and rose substantially in the land workers generally had statistically higher expos- last 2 time periods (AM   =  1.17 GDU (1.15, 1.19 ures than did other water workers, who generally had TP4 Downloaded from https://academic.oup.com/annweh/article/66/Supplement_1/i234/6395048 by DeepDyve user on 18 July 2022 Annals of Work Exposures and Health, 2022, Vol. 66, No. S1 i241 Figure 2. Modeled dermal exposure estimates of tar components (total hydrocarbons, benzene, ethylbenzene, toluene, xylene, n-hexane, and polycyclic aromatic hydrocarbons by broad activity groups by time period). Note that the scales (in GDUs) vary among the substances. ROVs = vessels piloting remotely operated vehicles. GDU = GuLF DREAM unit. statistically higher exposures than the rig, ROV, burner fire significant, the differences in the AMs for these workers control and RV workers, whether for oil or for tar (Fig. 1 were low and may not be meaningful. and 2 and SM, Table S6). Although often statistically Downloaded from https://academic.oup.com/annweh/article/66/Supplement_1/i234/6395048 by DeepDyve user on 18 July 2022 i242 Annals of Work Exposures and Health, 2022, Vol. 66, No. S1 For BTEX-H, the activities generally associated with PAHs, in contrast, had the highest maximums of ~219 the minimum values (<0.02 GDUs for all five chemicals) and ~588 GDUs for oil and tar, respectively. The broad varied (SM, Table S6). Benzene AMs for oil and for tar ranges of estimates for the exposures suggest that there across study participants reached 12.77 and 3.69 GDUs, should be sufficient contrast to differentiate between low respectively. AMs for ethylbenzene rose to 12.17 GDUs and high exposed individuals should these levels be asso- for oil and to 11.65 GDUs for tar. Similarly, the max- ciated with a health outcome. imum of the toluene AMs from oil was 17.45 GDUs; for There were statistically significant differences in the tar, the values rose to 42.37 GDUs. The maximums for AM exposures to THC in oil over time for the other xylene were 36.77 GDUs for oil and 88.18 GDUs for tar. water workers after TP1b and for land workers after The maximum AMs for n-hexane in oil were 2.22 GDUs TP2. Similarly, for tar, other water workers’ THC AMs and in tar, 5.56 GDUs. For PAHs, the maximum values rose after TP3, and land worker’s THC AMs rose after were 219.31 GDUs for oil and 587.98 GDUs for tar. TP4. We also evaluated THC exposure differences over Workers on the rig vessels, the ROV vessels and the broad groups of workers. Land workers had statistically burner fire control vessels were not considered exposed higher THC AM exposures from oil than did all other to dispersants (Table 2). The two components evaluated workers in TP1a to TP4. A similar pattern was seen for for dispersants were THC and xylene to represent pet- tar: land workers had higher THC AMs from tar in TP3, roleum distillates, hydrotreated light. The AM of the TP4, TP5, and TP6 than other water workers. The fact estimates for participants on land for THC were 25.41 that we found some differences suggests that we were and 18.88 GDUs and for xylene, 0.21 and 0.36 GDUs able to, at least to some degree, increase the accuracy of in TP1a and TP1b, respectively, the only time periods our estimates by considering jobs/activities/tasks, loca- for which dispersants were used. For participants on tion and time, thereby reducing misclassification error the water the AMs were for THC, 80.37 and for xylene, within the study participants. 1.39 GDUs, respectively. It seems counterintuitive that exposures to oil and Correlations among the substances in oil across all (for land workers) to tar should have risen over time. time periods combined were generally high (Table 1). The model is complex, however, and has many variables All substances had correlations of ρ ≥ 0.9 with each that affect the estimates, making it difficult to identify other except n-hexane. The n-hexane correlation with the single reason for this finding. In the online SM, we both THC and toluene was ρ = 0.87, with ethylbenzene, explore the effect of concentration, vapor pressure and rho = 0.84, xylene, ρ = 0.83 and with PAHs, ρ = 0.82. viscosity to learn how these varied by substance from Correlations in tar remained high for (ρ > 0.9) for the 25% weathering (around TP2, generally reflecting lower relationships between toluene, xylene, n-hexane and exposures from oil) to 40% weathering (TP5 and 6, re- PAHs. The relationships of THC and benzene, however, flecting higher exposures to tar). Viscosity appeared to with each other and with the other substances of interest be the predominant variable for ethylbenzene, toluene, were lower (ρ = 0.28–0.86 for THC, ρ = 0.70–0.89 for xylene and PAHs. Viscosity results in stickiness and the benzene). stickier the substance, the longer it will stay on the skin, thus increasing exposure. VP was the most important variable for THC and n-hexane. For both substances, Discussion the VP decreased substantially as the oil became more We describe procedures for the development of dermal weathered, resulting in an increase in the VP score (the exposure estimates derived from an updated and en- score being the inverse of VP) seen only with THC, ben- hanced version of DREAM (Gorman Ng et al., 2021). zene, and n-hexane, and therefore an increase in the We used participants’ interview responses when avail- exposure for THC and n-hexane. This decrease in VP able and JEM or industrial hygienist-entered values and the increase in viscosity should have resulted in an when not available. Even after deleting values <0.01 increase in benzene exposures over time; however, con- GDUs, minimum values were <0.05 GDUs for all sub- centration dominated the change seen in benzene by stances in oil or tar. For THC, values ranged up to ~5 being the second largest decrease among the substances. GDUs for oil and ~142 GDUs for tar. n-Hexane had See Stenzel, Arnold et  al. (2021) for more details on GDU levels of the same order of magnitude as those of weathering and its impact on the oil components. Less THC from oil, whereas the estimates for benzene, ethyl- of a contrast between benzene, THC, and n-hexane benzene, toluene, and xylene were substantially higher may have been seen had the differences in the VP score than THC for oil (up to 36.77 GDUs for xylene). For categories more closely reflected the differences in the VP tar, exposures were lower for the BTEX-H chemicals. (i.e. for 0–40% weathering, THC VP changed from 3986 Downloaded from https://academic.oup.com/annweh/article/66/Supplement_1/i234/6395048 by DeepDyve user on 18 July 2022 Annals of Work Exposures and Health, 2022, Vol. 66, No. S1 i243 Table 1. Correlations (ρ) between total hydrocarbons, benzene, ethylbenzene, toluene, xylene, n-hexane and polycyclic aromatic hydrocarbons in oil and tar. Correlation (ρ) among oil-related components (N = 1643 ) THC Benzene Ethylbenzene Toluene Xylene n-Hexane PAHs THC 1.00 0.98 >0.99 >0.99 0.98 0.87 0.98 Benzene 0.98 1.00 0.97 0.97 0.95 0.93 0.95 Ethylbenzene 1.00 >0.99 0.98 0.84 0.98 Toluene 1.00 0.98 0.87 0.98 Xylene 1.00 0.83 >0.99 n-Hexane 1.00 0.82 PAHs 1.00 Correlation (ρ) among tar-related components (N = 773) THC Benzene Ethylbenzene Toluene Xylene n-Hexane PAHs THC 1.00 0.28 0.77 0.68 0.77 0.86 0.80 Benzene 1.00 0.83 0.89 0.82 0.70 0.79 Ethylbenzene 1.00 >0.99 >0.99 0.97 >0.99 Toluene 1.00 >0.99 0.94 0.98 Xylene 1.00 0.96 >0.99 n-Hexane 1.00 0.97 PAHs 1.00 THC: total hydrocarbons. PAHs: polycyclic aromatic hydrocarbons. N = number of exposure groups. Table 2. Estimates of dispersant exposures (in GDUs). Chemical Broad group TP1a TP1b AM (LCL, UCL) AM (LCL, UCL) THC All land 25.41 (22.32, 28.93) 18.88 (17.26, 20.64) THC All other water operations NA 80.37 (62.54, 103.30) Xylene All land 0.21 (0.19, 0.24) 0.36 (0.33, 0.38) Xylene All other water operations NA 1.39 (1.08, 1.79) GDUs: GuLF DREAM units (see text for definition.) TP1a = time period 1a (April 22–May 14, 2010). TP1b = time period 1b (May 15–July 15, 2010). AM (AMLC, AMUC) = arithmetic mean (AM lower confidence value, AM upper confidence value). THC = total hydrocarbons. Workers on rig vessels, vessels piloting remotely operated vehicles (ROVs), burner fire control vessels, and research vessels were not identified as having dispersant exposures. NA = not applicable. No water workers in TP1a reported having contact with dispersants. to 821 Pa, respectively, whereas the scores changed from TP2 and in TP5 versus the median for activities only 0.01 to 0.1, respectively. For n-hexane, the respective performed in TP2 and found a higher median THC ex- values were 1025 to 36 Pa, with respective scores of 0.01 posure in the former (i.e. the first comparison of TP2 and to 1.0. For benzene, in contrast, those same respective TP5). In addition, the later activities were likely more as- values were 115 to 29 Pa but scores of 0.1 to 1.) sociated with tar (with the higher viscosity) than with oil A second reason for the rise in exposures is likely an (with the lower viscosity). artifact of the data, in that during the later time periods, The information used as inputs to the model was higher exposed activities comprise a larger percentage primarily from self-reports of the study participants for of the activities being performed. We compared the TP2 several reasons. We were able to observe only a limited median THC exposures of the activities performed in number of operations, as almost all OSRC activities Downloaded from https://academic.oup.com/annweh/article/66/Supplement_1/i234/6395048 by DeepDyve user on 18 July 2022 i244 Annals of Work Exposures and Health, 2022, Vol. 66, No. S1 had ceased by the time the GuLF STUDY was initiated. in the past. It is likely that reporting of activities in our In addition, there were thousands of workers across 4 study would be easier to recall than job classifications, states, and the oil was detected over more than 112 100 particularly with the short time interval between the spill 2 2 km (69  656 mi ) of water (Westerholm and Rauch, and the enrollment interview 1–3 years later and the un- 2016). Moreover, although there was guidance on what usual circumstances of the event. protective equipment to wear for some of the activities, We did not ask about changes over time. We know ac- on-site industrial hygienists were responsible for modi- tivities changed, as did the weathering of oil. While we fying the recommended equipment given the specific made allowances for changes in weathering, we do not work conditions encountered on any given day. know if the specific tasks performed for a particular ac- The estimates for the various oil-related substances tivity changed over time and had no way to adjust for generally were highly correlated (ρ > 0.9) for oil. In con- such an occurrence if it had transpired. It is unlikely how- trast, for tar the relationships between THC and ben- ever, that the respondents would have been able to differ- zene found lower correlations (ρ = 0.3–0.9), although entiate among exposure conditions in each of up to the the relationships among the other substances remained seven time periods of the study. Furthermore, participants high. The high correlations in general are not surprising, most often completed the enrollment questionnaire by however, because the variables associated with the re- telephone and the investigators were reluctant to add such ported job/activity/task were assigned the same weights detail to an already long interview. To reduce the potential regardless of the substance. The differences only came misclassification associated with not obtaining participant with substance-related variables, i.e. the concentration, information for the relevant time periods, we imputed vapor pressure score, and viscosity score. This finding missing data based on the responses from only those indi- may make it difficult to identify if a particular substance viduals who worked in each specific time period. This ap- is uniquely associated with an adverse health effect. proach could have biased estimates in other time periods It is not known how accurate the participants’ re- if participants tended to report the characteristics of the ports were. First, we asked about skin/clothing contact period with the greatest, or the least, skin burden. with tar, defined in the interview as a “solid or gooey Estimates for some of the variables were assigned oily residue”, and then with oil (“oil or oily water”) to by the study industrial hygienists. Of these, some were reduce the likelihood of participants confusing oil with substance-specific data (concentration, vapor pressure tar. Some participants still, however, may have confused score, viscosity score) that changed with the degree of the two, particularly after the oil had undergone some weathering. Other variables were the percentages of the weathering. This confusion primarily would have af- body parts exposed and the emission pathway, both of fected participants who worked on land, although par- which we did not think that respondents would have been ticipants performing water activities after TP3 (August able to provide. We had no way to validate our assign- 10–September 1, 2010) could also have been confused. ments because the oil spill response and clean-up effort Depending on the substance, this confusion could result was largely completed by the time the exposure assess- in an over- or underestimation of exposures. Second, we ment started, and we were able to observe few workers asked about frequency of skin exposure in terms of the during the performance of their jobs. We reviewed, proportion of a day (none, less than half, about half, however, the extensive amount of documentation, par- more than half, all of it), but we asked about hand ex- ticularly on weathering (see the SM in Stenzel, Arnold posure in terms of hours and minutes. In the home visit, et al., 2021) and the substantial number of photographs we asked how often contact with “a chemical” occurred taken of OSRC workers to obtain information on work with each body part (<1 day per month, 1-4 days per activities. month, 1–5 days per week or almost every day), and we We did not evaluate the day-to-day variability of ex- assumed that the frequency on the body part would have posures experienced by an individual due to the lack of been the same for all exposures of interest, which likely day-to-day information. The model developed point esti- introduced some error in the exposure estimation. It is mates from over 50 exposure variables. Given the amount not clear what the magnitude would be and whether the of uncertainty in the self-reports, this additional uncer- error would result primarily in an over or underestima- tainty was unlikely to have provided useful information. tion of exposure. In contrast, reports of jobs/activities/ In our evaluation of the GuLF DREAM model, we tasks was likely to have been good. Teschke et al. (2002), were unable to validate our actual estimates with meas- found that reporting of job classifications generally had urements taken during the OSRC, and we were unable good agreement with records (Teschke et al., 2002), and to find any useful dermal measurements taken on oil recollection was better for recent jobs than jobs further spill workers. The only oil spill dermal measurements we Downloaded from https://academic.oup.com/annweh/article/66/Supplement_1/i234/6395048 by DeepDyve user on 18 July 2022 Annals of Work Exposures and Health, 2022, Vol. 66, No. S1 i245 found were from a National Institute for Occupational and high-exposed workers to distinguish between the Safety and Health (NIOSH) study of workers responding most heavily and least exposed type of workers. to the Exxon Valdez spill (Gorman et al., 1991), which reported that dermal exposure levels were higher prior Conclusions to the work shift than after the work shift. Thus, we used two published datasets: one, for oil, represented by Assessment of dermal exposures is difficult, and in this a study measuring heavy fuel oil exposures in a variety study, there were no dermal measurements made during of oil-using industries (Christopher et al. 2007, 2011) the response and clean-up operations. We used a deter- and the other, for tar, represented by a study of an as- ministic model with participant-reported and industrial phalt paving operation (Cavallari et al., 2012). Overall, hygienist-supplied data. Dermal exposures to seven sub- the evaluation demonstrated a moderate correlation stances contained in oil and tar and to dispersants were (ρ = 0.59) between GuLF DREAM exposure estimates estimated for each of the jobs/activities/tasks by location and hand wash and wipe measurements. This correlation and time period reported by the nearly 24 000 workers in was somewhat lower than the corresponding correlation the study. The range of dermal estimates was substantial calculated for DREAM (ρ = 0.78) (van Wendel de Joode, for many of the exposures of interest. This provides some 2005). This is not surprising because the DREAM valid- confidence that there may be enough contrast to identify ation by van Wendel de Joode (2005) involved exposure exposure-related differences in workers’ health, should assessors who had often directly observed the workers they exist; however, the high correlation among most of during the exposure measurements. GuLF DREAM was the exposures makes it unclear whether we will be able to evaluated with previously collected datasets of other in- identify a particular substance among the many assessed. vestigators, so that study investigators did not have the Dermal exposure assessment is still in its infancy and the opportunity to observe the measured workers directly. usefulness of deterministic models needs further evaluation. We were unable to evaluate the model for body parts other than hands due to insufficient measurements in the Supplementary data published studies. The original DREAM model found poorer correlation with other body parts. Supplementary data are available at Annals of Work Exposures Limitations of this work include dependence on the and Health online. study participants’ self-reports; our inability to observe the response and clean-up activities; lack of time-varying information on specific work; lack of information on Acknowledgments variability; and lack of an evaluation on parts of the We thank Wendy McDowell and Kaitlyn Rousch of McDowell body other than hands. In addition, the reports of dis- Safety and Health Services, Inc. and Matthew Curry, Braxton persants may have been overreported; positive responses Jackson, John McGrath and Kate Christenbury of Social & Scientific suggest more participants had exposure than expected. Systems, Inc. for the tremendous help they provided on this study. Finally, we estimated exposure using a dimensionless We also thank the workers for their participation in this study. unit and do not know how it relates to absorption. This is, however, the first study to provide relative Funding dermal exposure estimates from working on the response and clean-up of an oil spill under the many varied activ- This study was funded by the NIH Common Fund and the ities that took place during the OSRC. We used a deter- Intramural Research Program of the National Institute of Health, National Institute of Environmental Health Sciences ministic model, an approach that has been used in other (ZO1 ES 102945). studies without measurement data (Vermeulen et  al., 2002). The evaluation, although limited, provides some information on both oil and asphalt (similar to tar) ex- Conflict of interest posures. The dermal estimates show a different exposure Prof Cherrie is currently undertaking consulting work related pattern from inhalation, allowing investigators to inves- to the Deepwater Horizon disaster. All of his involvement with tigate different routes of exposure and disease mechan- this paper was prior to any potential conflict of interest arising. isms. Finally, the estimates yielded large contrasts across the range of exposure levels for the substances of interest. This broad range, while likely to contain misclassification Data availability error, should allow investigators to examine risks associ- The data underlying this article will be shared on reasonable ated with categories of exposure, such as low-, medium-, request, consistent with protections for the privacy of study Downloaded from https://academic.oup.com/annweh/article/66/Supplement_1/i234/6395048 by DeepDyve user on 18 July 2022 i246 Annals of Work Exposures and Health, 2022, Vol. 66, No. S1 participants and existing multi-party agreements. Requests Deepwater Horizon oil spill clean-up operations. Ann Work should be made following instructions on the study website Expo Health; 66: i124–i139. https://gulfstudy.nih.gov. Kwok RK, Engel LS, Miller AK et al.; GuLF STUDY Research Team. (2017) The GuLF STUDY: a prospective study of per- sons involved in the deepwater horizon oil spill response References and clean-up. Environ Health Perspect; 125: 570–8. Arnold S, Stewart PA, Pratt GC et al. (2021) Estimation of aerosol NIOSH. (2011) NIOSH Deepwater Horizon Roster Summary TM Report. Available at https://www.cdc.gov/niosh/updates/ concentrations of oil dispersants COREXIT EC9527a and EC9500A during the Deepwater Horizon oil spill re- upd-12-19-11.html. (accessed 16 June 2018). NIOSH. (1994) Pocket Guide to Chemical Hazards. DHHS sponse and clean-up operations. Ann Work Expo Health; 66: i188–i201. (NIOSH) Publication No. 94-116. Pratt GC, Stenzel MR, Kwok RK et al. (2021) Modeled air pol- Cavallari JM, Osborn LV, Snawder JE et al. (2012) Predictors of dermal exposures to polycyclic aromatic compounds lution from in situ burning and flaring of oil and gas re- leased following the Deepwater Horizon disaster. Ann Work among hot-mix asphalt paving workers. Ann Occup Hyg; 56: 125–37. Expo Health; 66: i172–i187. Ramachandran G, Groth CP, Huynh TB et  al. (2021) Using Christopher  Y, van  Tongeren  M, Cowie  H et  al. (2007) Occupational dermal exposure to heavy fuel oils. Research real-time area VOC measurements to estimate total hydro- carbons exposures to workers involved in the Deepwater Report TM/07/05. Edinburgh, UK: IOM. Christopher Y, Van Tongeren M, Urbanus J et al. (2011) An as- Horizon oil spill. Ann Work Expo Health; 66: i156–i171. Stenzel MR, Arnold SF, Ramachandran G et al. (2021) Estimation sessment of dermal exposure to heavy fuel oil (HFO) in oc- TM cupational settings. Ann Occup Hyg; 55: 319–28. of airborne concentrations of oil dispersants COREXIT EC9527A and EC9500A, volatile components associated Gorman RW, Berardinelli SP, Bender TR. (1991) Exxon Valdez Alaska Oil Spill (1991) HETA 89-200 & 89-273-2111. with the Deepwater Horizon oil spill response and clean-up operations. Ann Work Expo Health; 66: i202–i217. Cincinnati, OH: National Institute for Occupational Safety and Health. Stenzel MR, Groth CP, Huynh TB et al. (2021) Exposure group development in support of the NIEHS GuLF STUDY. Ann Gorman  Ng  M, Cherrie  JW, Sleeuwenhoek  A et  al. (2021). GuLF DREAM: a model to estimate dermal exposure Work Expo Health; 66: i23–i55. Stewart P, Groth C, Huynh TB et al. (2021) Assessing expos- among oil spill response and clean-up workers. Ann Work Expo Health; 66: i218–i233. ures from the Deepwater Horizon oil spill response and clean-up. Ann Work Expo Health; 66; i3–i22. Groth CP, Banerjee S, Ramachandran G et al. (2021) Methods for the analysis of 26 million VOC area measurements Teschke K, Olshan AF, Daniels JL et al. (2002) Occupational ex- posure assessment in case–control studies: opportunities for during the Deepwater Horizon oil spill clean-up. Ann Work Expo Health; 66: i140–i155. improvement. Occup Environ Med; 59: 575–594. Van Wendel de Joode B, Brouwer DH, Vermeulen R et al. (2003) Groth CP, Huynh TB, Banerjee S et al. (2021) Linear relationships between total hydrocarbons and benzene, toluene, ethylben- DREAM: a method for semi-quantitative dermal exposure assessment. Ann Occup Hyg; 47: 71–87. zene, xylene, and n-hexane during the Deepwater Horizon response and clean-up. Ann Work Expo Health; 66: i71–i88. Van Wendel de Joode B, Vermeulen R, van Hemmen JJ et al. (2005) Accuracy of a semiquantitative method for Dermal Exposure Huynh TB, Groth CP, Ramachandran G et al. (2021a) Estimates of occupational inhalation exposures to six oil-related com- Assessment (DREAM). Occup Environ Med; 62: 623–32. Vermeulen R, Stewart P, Kromhout H. (2002) Dermal exposure pounds on the four rig vessels responding to the Deepwater Horizon oil spill. Ann Work Expo Health; 66: i89–i110. assessment in occupational epidemiologic research. Scand J Work Environ Health; 28: 371–85. Huynh TB, Groth CP, Ramachandran G et al. (2021b) Estimates of inhalation exposures to oil-related components on the Westerholm  DA, Rauch  SD III. (2016) Deepwater Horizon oil spill: final programmatic damage assessment and res- supporting vessels during the Deepwater Horizon oil spill. Ann Work Expo Health; 66: i111–i123. toration plan and final programmatic environmental im- pact statement. https://repository.library.noaa.gov/view/ Huynh TB, Groth CP, Ramachandran G et al. (2021c) Estimates of inhalation exposures among land workers during the noaa/18084 (accessed 6 June 2020)

Journal

Annals of Work Exposures and Health (formerly Annals Of Occupational Hygiene)Oxford University Press

Published: Apr 7, 2022

There are no references for this article.